Arsenic trioxide preferentially binds to the ring finger protein PML: understanding target selection of the drug

Cao Kaiming , Yaping Sheng , Shihui Zheng , Siming Yuan , Guangming Huang and Yangzhong Liu *
CAS Key Laboratory of Soft Matter Chemistry, CAS High Magnetic Field Laboratory, Department of Chemistry, University of Science and Technology of China, Hefei, Anhui 230026, China. E-mail:

Received 25th July 2018 , Accepted 18th September 2018

First published on 19th September 2018

Arsenic trioxide (ATO) is used in the clinic for the treatment of acute promyelocytic leukemia by targeting the protein PML. However, many zinc-finger proteins could also be reactive to arsenic in cells. Here we found that ATO preferentially binds to the ring-finger domain of PML in a protein mixture with zinc-finger domains. These results provide the molecular basis of the target selection of ATO in cells.

Significance to metallomics

Arsenic trioxide (ATO) is the most effective drug for the treatment of acute promyelocytic leukemia (APL). The protein PML is the drug target of ATO; however, many zinc-finger proteins (ZFPs) are also reactive to ATO. Here we found that ATO preferentially binds to PML in a protein mixture with ZFPs, showing selectivity of the drug. Arsenic can even transfer from As-bound ZFPs to PML. This work reveals that PML is the primary drug target of ATO, while the non-specific reactions with ZFPs could be associated with the side-effects of ATO.

Arsenic trioxide (ATO, As2O3) is the most effective drug for the treatment of acute promyelocytic leukemia (APL) in the clinic.1–3 However, arsenic is also a well-known carcinogenic element that is associated with a number of cancers, including skin, kidney, liver and bladder cancers.4–6 Mechanistic studies indicate that PML is the target of ATO in the treatment of APL.7,8 The fusion of PML-RARα is the main cause of APL, while ATO reacts with the ring finger domain of PML, resulting in the SUMOylation and degradation of the PML-RARα fusion protein.9,10 On the other hand, the carcinogenic property of arsenic is probably due to reaction with the zinc-finger domains of DNA repair proteins,11–14 and the failure of DNA repair could increase cancer risk and its development.15–17 ATO can also react with other zinc-proteins, such as cullin 3-RING box 1 (Cul3-Rbx1),18 estrogen receptor-α (ERα),19 metal-activated transcription factor 1 (MTF1),20 and ten-eleven translocation (Tet).21 These reactions could also be involved in the biological function of ATO. These results highlight the significance of zinc-binding proteins in the mechanisms of arsenic action. Therefore, the target selection of arsenite in cells is crucial for the drug efficacy and side-effects of ATO.

The reactivity of zinc-finger proteins to arsenic is highly dependent on the coordination residues of zinc binding sites.22–24 Kinetic and thermodynamic studies showed that arsenite, a solution of ATO, preferentially binds to the C3H1 and C4 types of zinc-finger proteins.22–24 Since both zinc-finger domains and ring-finger domains contain a similar zinc-coordination, it is interesting to know whether ATO exhibits selectivity between two types of proteins. This result could provide information for understanding the target selection of ATO in cells.

In this work, we have investigated the arsenite reactions of the ring-finger domain of PML (aa 49–104), which contains C3H1 and C4 zinc binding motifs. By comparison, the reactions of the zinc-finger protein NCp7 (aa 12–55, containing two C3H1 motifs) and ERα (aa 180–241, containing two C4 motifs) were conducted and their competition for arsenite was investigated. The proteins were obtained by overexpression in E. coli according to the literature (see the ESI, Fig. S1 and S2).23

The reactions of proteins with arsenite were analyzed using different spectroscopic methods in this work. Both PML and NCp7 exhibit intrinsic fluorescence with excitation at 280 nm. Adding arsenite quenched the fluorescence of PML and NCp7, indicating that both proteins are able to react with arsenite (Fig. S3A, ESI). UV measurements showed that adding arsenite to PML and NCp7 led to increased absorption in the range of 240–320 nm, suggesting the formation of S–As bonds (Fig. S4, ESI). The binding of As(III) to the thiol group of cysteine residues was also confirmed by the zinc-release assay (see below), which is in agreement with the literature report.9 The reaction of cysteine residues at zinc-binding sites typically interferes with zinc coordination and protein folding. CD measurements confirmed that arsenite reactions disrupt the secondary structure of PML and NCp7 (Fig. S3B, ESI). This result is consistent with the literature report.9,23

The reactions of arsenite with proteins were further analyzed using ESI-MS measurements. In the reaction of PML, two major peaks were observed, indicating the formation of mono- and bi-As binding adducts (Fig. 1A). Only a mono-As binding adduct was observed on NCp7 under the same reaction conditions (Fig. 1B). In addition, the As-free signals were observed only for NCp7, but not for PML. This result implies that arsenite prefers to react with PML over NCp7 between two proteins. This assumption can be confirmed by 2D 1H–15N HSQC NMR spectra. The spectra were recorded on 15N-labeled proteins, and the chemical shifts are well in agreement with the literature.9,23 The reaction of 2 molar equivalents of arsenite clearly perturbed the spectrum of PML, while only a few peaks changed in the reaction of NCp7 under the same conditions. The large perturbation of NCp7 signals was observed when 10 molar equivalents of arsenite were added (Fig. S5, ESI), indicating that NCp7 is less reactive than PML to arsenite.

image file: c8mt00202a-f1.tif
Fig. 1 Reactions of arsenite with PML and NCp7. (A and B) ESI-MS spectra of analysis of the adducts from the arsenite reaction of PML (A) or NCp7 (B). The reactions were performed on 10 μM proteins and 30 μM arsenite at 37 °C for 2 hours. The selected regions of +4 charged peaks of PML and +5 charged peaks of NCp7 are shown. (C and D) 1H–15N HSQC NMR spectra of NCp7 (C) or PML (D) reacting with 2 molar equivalents of arsenite for 2 h. Superposition of the spectra of the proteins before (blue) and after (red) reaction with arsenite in 20 mM phosphate buffer (pH 5.8) containing 100 mM NaCl.

To understand the competition between PML and NCp7 in the arsenite reaction, we directly analyzed the reaction in a mixture of the two proteins. This competition reaction can also avoid any possible differences under solution conditions that might influence the reaction of the two proteins.24 By taking advantage of the atomic resolution of NMR spectroscopy, signals of each protein can clearly be distinguished in the mixture sample. Results show that the 2D 1H–15N HSQC NMR spectrum recorded on the mixture of PML and NCp7 (Fig. 2A) is nearly identical to the overlap of two spectra of the single proteins (Fig. 2B). Therefore, signals of each protein can easily be identified by the superposition of the spectrum of single PML or NCp7 (Fig. S6, ESI). Adding 5 molar equivalents of arsenite to the protein mixture caused a significant change in some peaks, while many other peaks remain unchanged (Fig. 2C). The circles on the map clearly indicate that the signals changed after the As(III) reaction. By comparing with the spectra of the two proteins, all these signals can be identified from the PML protein, whereas those signals of NCp7 were not changed in the reaction (Fig. S7, ESI). Moreover, arsenite titration was performed on this protein mixture and the HSQC NMR spectra were recorded (Fig. S8, ESI). Ten representative peaks (all of them finally disappeared at high arsenic concentration) of each protein were chosen and their intensity during the titration was measured. Data show that signals of PML decrease significantly with the concentration of arsenite, while only little change occurred on NCp7 (Fig. 2D). This result confirmed that PML is more reactive than NCp7 to arsenite.

image file: c8mt00202a-f2.tif
Fig. 2 1H–15N HSQC NMR analyses of the reaction of arsenite with the mixture of PML and NCp7. (A) Spectrum of the mixture of two proteins (PML + NCp7). (B) Superposition of the spectra of PML (red) and NCp7 (green). (C) Superposition of the spectra of the protein mixture (PML + NCp7) before (blue) and after (red) reaction with arsenite (5 equiv. for 2 h). All spectra were recorded at 298 K on 0.2 mM proteins in 20 mM phosphate buffer (pH 5.8) containing 100 mM NaCl. (D) The signal intensity changes with the ratio of [As]/[Protein]. The peak intensity was the average of 10 representative peaks selected from each protein on the HSQC spectra of the mixture of PML and NCp7. Error bars denote the standard deviations.

The competition of PML and NCp7 in the reaction of arsenite was also investigated using ESI-MS measurements. The signals of PML and NCp7 can be well distinguished in the protein mixture due to their different molecular weights. The binding of As(III) to proteins can be observed upon incubation with different amounts of arsenite. In the reaction of equimolar arsenite, both mono- and bi-arsenic binding adducts were largely formed on the protein PML (As–PML and As2–PML), whereas no NCp7 adducts were generated (Fig. 3 and Table S2, ESI). This result confirmed that arsenite preferentially reacts with PML. The reaction of NCp7 occurred only in the presence of excess arsenite. With five or ten molar equivalents of arsenite, PML completely reacted and the bi-As adduct became dominant on the spectrum, and then the NCp7 adducts can be observed. In this case there is still some unreacted NCp7 detected. This result is in agreement with NMR spectra that PML is more reactive than NCp7 to arsenite.

image file: c8mt00202a-f3.tif
Fig. 3 ESI-MS analysis of arsenite reactions with PML and NCp7 together. The reactions were performed on a mixture of 10 μM PML and 10 μM NCp7 with different amounts of arsenite (1, 5 and 10 molar equivalents) at 37 °C for 2 hours. Selected regions show the +5 charged peaks of NCp7 and the +4 charged peaks of PML. Asterisks denote the addition of Na+ on the protein.

As PML also contains a C4 zinc-coordination motif, the competition of PML and ERα in the reaction of arsenite was also investigated. The result showed that mono- and bi-As binding adducts were largely formed on the protein PML in the presence of 1.5 molar equivalents of arsenite, while no ERα adducts were detected (Fig. S9, ESI). This result reveals that the ring-finger of PML is more reactive than both NCp7 (C3H1 zinc-finger) and ERα (C4 zinc-finger) to arsenite.

Since the non-specific As(III) binding to other proteins could occur in cells, we tested whether the As-bound protein can transfer arsenic to PML. The reaction was performed on PML and As-bound NCp7. The 15N-labeled PML and natural isotopic abundance As-NCp7 proteins were used in the reaction, so that only signals of PML can be detected on the 2D 1H–15N HSQC NMR spectra. The reaction with As-NCp7 significantly perturbed the spectrum of PML, and a large number of signals shifted (Fig. 4A). The spectra alteration is similar to the reaction of free arsenite, showing the As(III) binding to PML. This result indicates that arsenic can transfer from NCp7 to PML. The reaction was also conducted on the 15N-labeled NCp7 and natural isotopic abundance As-PML protein, in which only signals of NCp7 can be detected on the NMR spectra. On the contrary to the previous reaction, no change in the spectra was observed (Fig. 4B). This result indicates that PML can take As(III) from NCp7, but not vice versa.

image file: c8mt00202a-f4.tif
Fig. 4 (A) Superposition of 1H–15N HSQC NMR spectra of 15N-labeled PML before (blue) and after (red) incubation with 5 molar equivalents of natural isotopic abundance As-NCp7. As-NCp7 was prepared by the apo-NCp7 reaction with arsenite. (B) Superposition of NMR spectra of 15N-labeled NCp7 before (blue) and after (red) incubation with 5 molar equivalents of natural isotopic abundance As-PML. As-PML was prepared by the apo-PML reaction with arsenite. NMR spectra were recorded at 298 K in 20 mM phosphate buffer (pH 5.8) containing 100 mM NaCl.

It has been well investigated that the arsenic reactivity of zinc-finger proteins is dependent on the type of zinc-binding residues.22 Hence, it is interesting to understand the reason why arsenite demonstrated a high selectivity to PML over NCp7 and ERα. First, we measured the binding affinity of As(III) to two proteins by fluorescence titration. The reactions of arsenite with apo-proteins lead to fluorescence quenching (Fig. S10 and S12A, ESI), therefore, the apparent dissociation constants (Kd) were obtained by fitting the titration curves (Fig. 5A). Data indicate that As(III) has a similar binding affinity to NCp7 (Kd = 3.47 ± 0.30 μM), ERα (Kd = 3.20 ± 0.28 μM) and PML (Kd = 2.31 ± 0.17 μM). This result suggests that the selectivity of arsenite is not due to the As(III) binding affinity.

image file: c8mt00202a-f5.tif
Fig. 5 (A) Fluorescence of PML and NCp7 quenched by arsenite. 20 μM proteins were incubated with arsenite at 37 °C for 30 min in the presence of 20 mM phosphate buffer, pH 5.8. (B) UV absorbance of Zn(PAR)2 at 500 nm in the reaction with PML and NCp7. Proteins were titrated to 15 μM Zn(PAR)2 in 20 mM phosphate buffer in the presence of 20 μM additional PAR to eliminate free Zn2+ ions. All reactions were performed at 37 °C. The color denotes the protein PML (red) or NCp7 (black).

Next, we measured the Zn(II) binding affinity to three proteins. Since the As(III) coordination leads to zinc-ejection from proteins, the easy zinc release should cause the protein to more readily bind arsenic.25 The assay was performed using a zinc dye PAR, as the formation of the [Zn(PAR)2] complex exhibits a characteristic absorption of 500 nm. Adding apo-proteins to [Zn(PAR)2] reduced the absorbance at 500 nm (Fig. S11 and S12B, ESI), showing that three proteins are able to compete with PAR in zinc coordination. The concentration dependent measurements show that NCp7 and ERα can take Zn(II) from [Zn(PAR)2] more efficiently than PML (Fig. 5B and Fig. S12B, ESI). This result indicates that NCp7 and ERα possess a higher Zn(II) binding affinity than PML. Therefore, the lower Zn(II) binding affinity of PML enables this protein to more readily bind As(III), and results in the higher selectivity of PML in the arsenite reaction.

In addition to thermodynamic preference, the kinetic activity could also influence the target selection of metallodrugs.26,27 Hence a time dependent fluorescence measurement was conducted for the arsenite reactions (Fig. S13A, ESI). Results showed that the reaction of NCp7 is faster than PML; the half-life (t1/2) of NCp7 (1.2 min) is obviously shorter than that of PML (3.6 min) (Fig. S13A, ESI). Zinc-release assays also showed that the reaction of arsenite caused more rapid zinc ejection from NCp7 than from PML (Fig. S13B, ESI). Nevertheless, the reaction rates of both proteins are in a short minutes range, indicating that they are kinetically competitive. More importantly, the above mentioned arsenic transfer assay indicates that As(III) can transfer from NCp7 to PML even if it reacts with NCp7 in advance. Therefore, the kinetic difference would not be a big issue in target selection of arsenite. Competition assays confirmed this conclusion as the arsenite reaction occurs on PML prior to NCp7.

Target selection determines the medicinal application of metallodrugs since their coordinations are largely non-specific.28,29 Although PML is the proposed drug target of ATO, As(III) tends to bind to any proteins containing multi-cysteine coordination sites.19,30–32 Previous studies indicate that ATO selectively reacts with zinc-finger proteins with three or more cysteine residues (C3H1 or C4 type) since As(III) prefers thiol binding and forms a pyramidal AsS3 coordination.23 Similar to zinc-finger proteins, PML contains C3H1 and C4 zinc-binding centers.10 Therefore, the reactivity of ATO to PML and zinc-finger proteins is crucial for the target selection of the drug. Here in this work, we found that the ring-finger of PML exhibits a higher reactivity than the zinc-finger of NCp7 (C3H1) and ERα (C4) in the arsenite reaction (Scheme 1). This result indicates that PML has high priority in the arsenic reaction in the presence of other reactive proteins, which is crucial for the medicinal application of ATO. Although different arsenite concentrations were used in various assays in this work, the cellular conditions could further enhance the arsenite reaction with proteins.24 On the other hand, the toxicity and side-effects of ATO are probably associated with the off-target binding, as minor binding of zinc-finger proteins is observed in a concentration dependent manner.

image file: c8mt00202a-s1.tif
Scheme 1 Illustration of the selectivity of ATO in reactions of PML and zinc-finger proteins.


In summary, we have investigated the reaction of arsenic trioxide (ATO), an anti-APL drug, with the target protein PML. Results indicate that arsenite exhibits a high selectivity to the ring-finger domain of PML over the zinc-finger domain of NCp7 and ERα. The 2D NMR and ESI-MS measurements showed that arsenite reacts with PML more efficiently than with NCp7. The arsenite reaction in the mixture of two proteins revealed that As(III) is able to bind to PML prior to NCp7. The arsenic binding leads to the zinc-ejection from proteins and disrupts the protein folding. In addition, arsenic can transfer from As-bound NCp7 to PML, even if it reacts with NCp7 in advance. However, the reverse arsenic transfer cannot happen. These results indicate that the ring-finger of PML is capable of competing with zinc-finger proteins in As(III) binding. Metal binding assays show that the three proteins have similar As(III) binding affinities, however, NCp7 and ERα demonstrated a higher Zn(II) binding affinity than PML. This difference makes PML more readily react with arsenite, resulting in the selectivity of the protein. These in vitro reaction results reveal that PML is the primary drug target of ATO even in the presence of other zinc-finger proteins. This work provides a novel insight into the molecular mechanism of the drug ATO in terms of target selection.

Conflicts of interest

There are no conflicts to declare.


This work was supported by the National Key R&D Program of China (2017YFA0505400), the National Science Foundation of China (21877103 and 21573213), the Major Program of Development Foundation of Hefei Center for Physical Science and Technology (2018ZYFX004), the Suzhou Science and Technology Project (SYG201624), the Jiangsu Natural Science Foundation (BK20151238) and the Collaborative Innovation Center of Suzhou Nano Science and Technology.

Notes and references

  1. S. Kayser, J. Krzykalla, M. A. Elliott, K. Norsworthy, P. Gonzales, R. K. Hills, M. R. Baer, Z. Racil, J. Mayer, J. Novak, P. Zak, T. Szotkowski, D. Grimwade, N. H. Russell, R. B. Walter, E. H. Estey, J. Westermann, M. Gorner, A. Benner, A. Kramer, B. D. Smith, A. K. Burnett, C. Thiede, C. Rollig, A. D. Ho, G. Ehninger, R. F. Schlenk, M. S. Tallman, M. J. Levis and U. Platzbecker, Characteristics and outcome of patients with therapy-related acute promyelocytic leukemia front-line treated with or without arsenic trioxide, Leukemia, 2017, 31, 2347–2354 CrossRef CAS PubMed.
  2. U. Platzbecker, G. Avvisati, L. Cicconi, C. Thiede, F. Paoloni, M. Vignetti, F. Ferrara, M. Divona, F. Albano, F. Efficace, P. Fazi, M. Sborgia, E. Di Bona, M. Breccia, E. Borlenghi, R. Cairoli, A. Rambaldi, L. Melillo, G. La Nasa, W. Fiedler, P. Brossart, B. Hertenstein, H. R. Salih, M. Wattad, M. Lubbert, C. H. Brandts, M. Hanel, C. Rollig, N. Schmitz, H. Link, C. Frairia, E. M. Pogliani, C. Fozza, A. M. D'Arco, N. Di Renzo, A. Cortelezzi, F. Fabbiano, K. Dohner, A. Ganser, H. Dohner, S. Amadori, F. Mandelli, G. Ehninger, R. F. Schlenk and F. Lo-Coco, Improved Outcomes With Retinoic Acid and Arsenic Trioxide Compared With Retinoic Acid and Chemotherapy in Non-High-Risk Acute Promyelocytic Leukemia: Final Results of the Randomized Italian-German APL0406 Trial, J. Clin. Oncol., 2017, 35, 605–612 CrossRef CAS PubMed.
  3. H. Gill, R. Yim, H. K. K. Lee, V. Mak, S. Y. Lin, B. Kho, S. F. Yip, J. S. M. Lau, W. Li, H. W. Ip, Y. Y. Hwang, T. S. Y. Chan, E. Tse, W. Y. Au, C. R. Kumana and Y. L. Kwong, Long-Term Outcome of Relapsed Acute Promyelocytic Leukemia Treated With Oral Arsenic Trioxide-Based Reinduction and Maintenance Regimens: A 15-Year Prospective Study, Cancer, 2018, 124, 2316–2326 CrossRef CAS PubMed.
  4. U. Schuhmacher-Wolz, H. H. Dieter, D. Klein and K. Schneider, Oral exposure to inorganic arsenic: evaluation of its carcinogenic and non-carcinogenic effects, Crit. Rev. Toxicol., 2009, 39, 271–298 CrossRef CAS PubMed.
  5. J. Bach, J. Peremarti, B. Annangi, R. Marcos and A. Hernandez, Oxidative DNA damage enhances the carcinogenic potential of in vitro chronic arsenic exposures, Arch. Toxicol., 2016, 90, 1893–1905 CrossRef CAS PubMed.
  6. E. Kamynina, E. R. Lachenauer, A. C. DiRisio, R. P. Liebenthal, M. S. Field and P. J. Stover, Arsenic trioxide targets MTHFD1 and SUMO-dependent nuclear de novo thymidylate biosynthesis, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, E2319–E2326 CrossRef CAS PubMed.
  7. V. Lallemand-Breitenbach, M. Jeanne, S. Benhenda, R. Nasr, M. Lei, L. Peres, J. Zhou, J. Zhu, B. Raught and H. de The, Arsenic degrades PML or PML - RAR alpha through a SUMO-triggered RNF4/ubiquitin-mediated pathway, Nat. Cell Biol., 2008, 10, 547–555 CrossRef CAS PubMed.
  8. B. Chen, Q. Liu, A. Popowich, S. Shen, X. Yan, Q. Zhang, X. F. Li, M. Weinfeld, W. R. Cullen and X. C. Le, Therapeutic and analytical applications of arsenic binding to proteins, Metallomics, 2015, 7, 39–55 RSC.
  9. X. Zhang, X. Yan, Z. Zhou, F. Yang, Z. Wu, H. Sun, W. Liang, A. Song, V. Lallemand-Breitenbach, M. Jeanne, Q. Zhang, H. Yang, Q. Huang, G. Zhou, J. Tong, Y. Zhang, J. Wu, H. Hu, H. de The, S. J. Chen and Z. Chen, Arsenic Trioxide Controls the Fate of the PML-RAR alpha Oncoprotein by Directly Binding PML, Science, 2010, 328, 240–243 CrossRef CAS PubMed.
  10. P. Wang, S. Benhenda, H. Wu, V. Lallemand-Breitenbach, T. Zhen, F. Jollivet, L. Peres, Y. Li, S. J. Chen, Z. Chen, H. de The and G. Meng, RING tetramerization is required for nuclear body biogenesis and PML sumoylation, Nat. Commun., 2018, 9, 1277 CrossRef PubMed.
  11. P. Mandal, Molecular insight of arsenic-induced carcinogenesis and its prevention, Naunyn-Schmiedeberg's Arch. Pharmacol., 2017, 390, 443–455 CrossRef CAS PubMed.
  12. K. Piatek, T. Schwerdtle, A. Hartwig and W. Bal, Monomethylarsonous acid destroys a tetrathiolate zinc finger much more efficiently than inorganic arsenite: Mechanistic considerations and consequences for DNA repair inhibition, Chem. Res. Toxicol., 2008, 21, 600–606 Search PubMed.
  13. H. Liu, Z. Zhang, X. Chi, Z. Zhao, D. Huang, J. Jin and J. Gao, Arsenite-loaded nanoparticles inhibit PARP-1 to overcome multidrug resistance in hepatocellular carcinoma cells, Sci. Rep., 2016, 6, 31009 Search PubMed.
  14. J. Huestis, X. Zhou, L. Chen, C. Feng, L. G. Hudson and K. Liu, Kinetics and thermodynamics of zinc(II) and arsenic(III) binding to XPA and PARP-1 zinc finger peptides, J. Inorg. Biochem., 2016, 163, 45–52 CrossRef CAS PubMed.
  15. W. Ding, W. Liu, K. L. Cooper, X. Qin, P. L. de Souza Bergo, L. G. Hudson and K. Liu, Inhibition of poly(ADP-ribose) polymerase-1 by arsenite interferes with repair of oxidative DNA damage, J. Biol. Chem., 2009, 284, 6809–6817 CrossRef CAS PubMed.
  16. F. Zhang, M. Paramasivam, Q. Cai, X. Dai, P. Wang, K. Lin, J. Song, M. M. Seidman and Y. Wang, Arsenite binds to the RING finger domains of RNF20-RNF40 histone E3 ubiquitin ligase and inhibits DNA double-strand break repair, J. Am. Chem. Soc., 2014, 136, 12884–12887 CrossRef CAS PubMed.
  17. N. Holcomb, M. Goswami, S. G. Han, T. Scott, J. D'Orazio, D. K. Orren, C. G. Gairola and I. Mellon, Inorganic arsenic inhibits the nucleotide excision repair pathway and reduces the expression of XPC, DNA Repair, 2017, 52, 70–80 CrossRef CAS PubMed.
  18. J. Jiang, L. M. Tam, P. Wang and Y. Wang, Arsenite Targets the RING Finger Domain of Rbx1 E3 Ubiquitin Ligase to Inhibit Proteasome-Mediated Degradation of Nrf2, Chem. Res. Toxicol., 2018, 31, 380–387 Search PubMed.
  19. A. Hussain, V. A. Raveendran, S. Kundu, T. Samanta, R. Shunmugam, D. Pal and J. D. Sarma, Mechanisms of Arsenic-Induced Toxicity with Special Emphasis on Arsenic-Binding Proteins, Arsenic-Analytical and Toxicological Studies, IntechOpen, 2018, pp. 57–80 Search PubMed.
  20. R. Ganger, R. Garla, B. P. Mohanty, M. P. Bansal and M. L. Garg, Protective Effects of Zinc Against Acute Arsenic Toxicity by Regulating Antioxidant Defense System and Cumulative Metallothionein Expression, Biol. Trace Elem. Res., 2016, 169, 218–229 CrossRef CAS PubMed.
  21. S. Liu, J. Jiang, L. Li, N. J. Amato, Z. Wang and Y. Wang, Arsenite Targets the Zinc Finger Domains of Tet Proteins and Inhibits Tet-Mediated Oxidation of 5-Methylcytosine, Environ. Sci. Technol., 2015, 49, 11923–11931 CrossRef CAS PubMed.
  22. X. Zhou, X. Sun, K. L. Cooper, F. Wang, K. Liu and L. G. Hudson, Arsenite Interacts Selectively with Zinc Finger Proteins Containing C3H1 or C4 Motifs, J. Biol. Chem., 2011, 286, 22855–22863 CrossRef CAS PubMed.
  23. L. Zhao, S. Chen, L. Jia, S. Shu, P. Zhu and Y. Liu, Selectivity of arsenite interaction with zinc finger proteins, Metallomics, 2012, 4, 988–994 RSC.
  24. L. Zhao, Z. Wang, Z. Xi, D. Xu, S. Chen and Y. Liu, The Reaction of Arsenite with Proteins Relies on Solution Conditions, Inorg. Chem., 2014, 53, 3054–3061 CrossRef CAS PubMed.
  25. G. Jiang, Z. Gong, X. F. Li, W. R. Cullen and X. C. Le, Interaction of trivalent arsenicals with metallothionein, Chem. Res. Toxicol., 2003, 16, 873–880 Search PubMed.
  26. L. Zhao, Q. Cheng, Z. Wang, Z. Xi, D. Xu and Y. Liu, Cisplatin binds to human copper chaperone Cox17: the mechanistic implication of drug delivery to mitochondria, Chem. Commun., 2014, 50, 2667–2669 RSC.
  27. S. Chen, D. Xu, H. Jiang, Z. Xi, P. Zhu and Y. Liu, Trans-Platinum/Thiazole Complex Interferes with Sp1 Zinc-Finger Protein, Angew. Chem., Int. Ed., 2012, 51, 12258–12262 CrossRef CAS PubMed.
  28. M. Wenzel and A. Casini, Mass spectrometry as a powerful tool to study therapeutic metallodrugs speciation mechanisms: Current frontiers and perspectives, Coord. Chem. Rev., 2017, 352, 432–460 CrossRef CAS.
  29. X. Yang, Regulating cellular stress responses: an emerging strategy for rational metallodrug design, Future Med. Chem., 2018, 10, 611–614 CrossRef CAS PubMed.
  30. S. Shen, X. F. Li, W. R. Cullen, M. Weinfeld and X. C. Le, Arsenic Binding to Proteins, Chem. Rev., 2013, 113, 7769–7792 CrossRef CAS PubMed.
  31. M. Lu, H. Wang, X. F. Li, L. L. Arnold, S. M. Cohen and X. C. Le, Binding of dimethylarsinous acid to Cys-13 alpha of rat hemoglobin is responsible for the retention of arsenic in rat blood, Chem. Res. Toxicol., 2007, 20, 27–37 Search PubMed.
  32. M. Lu, H. Wang, Z. Wang, X. F. Li and X. C. Le, Identification of reactive cysteines in a protein using arsenic labeling and collision-induced dissociation tandem mass spectrometry, J. Proteome Res., 2008, 7, 3080–3090 CrossRef CAS PubMed.


Electronic supplementary information (ESI) available: Experimental details, protein expression and purification, NMR spectroscopy, and reaction kinetics. See DOI: 10.1039/c8mt00202a

This journal is © The Royal Society of Chemistry 2018